Suche Bilder Maps Play YouTube News Gmail Drive Mehr »
Anmelden
Nutzer von Screenreadern: Klicke auf diesen Link, um die Bedienungshilfen zu aktivieren. Dieser Modus bietet die gleichen Grundfunktionen, funktioniert aber besser mit deinem Reader.

Patentsuche

  1. Erweiterte Patentsuche
VeröffentlichungsnummerUS20100293174 A1
PublikationstypAnmeldung
AnmeldenummerUS 12/464,496
Veröffentlichungsdatum18. Nov. 2010
Eingetragen12. Mai 2009
Prioritätsdatum12. Mai 2009
Auch veröffentlicht unterUS8719249
Veröffentlichungsnummer12464496, 464496, US 2010/0293174 A1, US 2010/293174 A1, US 20100293174 A1, US 20100293174A1, US 2010293174 A1, US 2010293174A1, US-A1-20100293174, US-A1-2010293174, US2010/0293174A1, US2010/293174A1, US20100293174 A1, US20100293174A1, US2010293174 A1, US2010293174A1
ErfinderPaul N. Bennett, David Maxwell Chickering, Kevyn B. Colllins-Thompson, Susan Dumais, Daniel J. Liebling
Ursprünglich BevollmächtigterMicrosoft Corporation
Zitat exportierenBiBTeX, EndNote, RefMan
Externe Links: USPTO, USPTO-Zuordnung, Espacenet
Query classification
US 20100293174 A1
Zusammenfassung
Techniques and systems are disclosed that provide for constructing a query classification index that can be used to classify a query into relevant categories. Where documents in an index are classified into one or more category predictions for a category hierarchy, classification metadata is generated for categories to which a document in the index has been classified. Further, the classification metadata is associated to the corresponding documents in the index. Additionally, a query of the index can be classified using the metadata associated to the documents in the index, and query results can be provided that are classified by the one or more categories identified by the classification of the query.
Bilder(8)
Previous page
Next page
Ansprüche(20)
1. A method for providing query results based on categorization of searched content, comprising:
generating classification metadata that identify a category to which a document in an index of documents is classified, where documents in the index have been classified into one or more category predictions for a category hierarchy; and
associating the classification metadata to the document in the index.
2. The method of claim 1, comprising:
classifying a query of the index using the metadata associated to the documents in the index; and
providing query results classified by one or more categories using the classification of the query.
3. The method of claim 1, classifying documents in an index into category predictions for a category hierarchy, comprising one or more of:
using a trained classifier to identify one or more categories for the document, based on a desired threshold; and
determining a retrieval relevance score to identify one or more categories for the document, based on a desired threshold.
4. The method of claim 2, classifying a query of the index using the metadata associated to the documents in the index, comprising:
identifying document results for the query from the index;
identifying one or more categories for respective documents from the results using a distribution of the categories for the results provided by the metadata; and
providing one or more categories for the query based on a distribution of the one or more identified categories for respective documents in the results.
5. The method of claim 1, comprising building a mapping of query terms to a category distribution for the index, comprising:
mapping respective terms in a document in the index to the one or more category predictions for the document;
aggregating the terms mapped for the index to create a distribution of category predictions for the respective terms.
6. The method of claim 5, classifying a query of the index using the metadata associated to the documents in the index, comprising:
mapping respective query terms to the distribution of category predictions for the respective terms in the index;
identifying one or more categories for the query using the query term mapping; and
providing one or more predicted relevant topic models for the query based on the distribution of the identified categories for the query.
7. The method of claim 2, comprising determining ambiguity for the query comprising using an ambiguity function that accounts for a spread of the distribution of the categories associated with the query.
8. The method of claim 7, determining an ambiguity score for the query comprising measuring the spread of the distribution of a desired number of categories associated with the query.
9. The method of claim 2, comprising determining a query similarity between the query and an alternate query comprising comparing the distribution of the categories associated with the query and the distribution of the categories associated with the alternate query.
10. The method of claim 9, determining a query similarity score for the query with respect to the alternate query comprising calculating a similarity function using the distribution of the categories associated with the query and the distribution of the categories associated with the alternate query.
11. The method of claim 2, comprising determining a query quality characteristic, comprising combining the determination of query ambiguity with the determination of query similarity between the query and an alternate query.
12. The method of claim 2, comprising determining a query refinement score that is used to measure whether an alternate query is one or more of:
a more specific refinement of one or more portions of the query;
a more general refinement of one or more portions of the query; and
an orthogonal concept to one or more portions of the query;
13. The method of claim 1, where a query comprises one or more of:
keyword terms from a search-based query;
keywords associated with an Internet-based advertisement;
tags terms associated with a file; and
keywords associated with a caption.
14. The method of claim 1, classifying an index of documents comprising classifying a federation of documents from different source collections.
15. The method of claim 1, comprising one or more of:
generating classification metadata at crawl time of a source of documents for the index; and
associating the classification metadata to the document at crawl time of a source of documents for the index.
16. A system for providing query results based on categorization of searched content, comprising:
computer-based storage configured to store classification metadata associated with documents in an index of documents;
a metadata generator configured to generate classification metadata that identify a category to which a document in the index has been classified, where documents in the index have been classified into one or more category predictions for a category hierarchy; and
a metadata association component, operably coupled to the storage, and configured to associate the classification metadata to the corresponding document in the index.
17. The system of claim 16, comprising:
a query classifier configured to classify a query of the index using the metadata associated to the documents in the index comprising a query distribution component configured to generate a distribution of categories, provided by the metadata, for query results; and
a classified query results provider operably coupled to the query classifier and configured to provide query results classified by one or more categories using the classification of the query from the query classifier.
18. The system of claim 16 comprising a trained classifier configured to identify one or more categories for a document in the index, based on a desired threshold.
19. The system of claim 16, comprising:
a document term mapper configured to:
map respective terms in a document in the index to the one or more category predictions for the document; and
aggregate the terms mapped for the index to create a distribution of category predictions for the respective terms; and
a mapped query classifier configured to:
map respective query terms, for a query to the index, to the distribution of category predictions for the respective terms in the index;
identify one or more categories for the query using the query term mapping; and
provide one or more predicted relevant topic models for the query based on the distribution of the identified categories for the query.
20. A method for providing query results based on categorization of searched content, comprising:
classifying document in an index into topic category prediction for a category hierarchy, comprising using a trained classifier to identify one or more categories for the document, based on a desired threshold;
generating classification metadata, at crawl time of a source of documents for the index, that identify a topic category to which a document has been classified;
associating the classification metadata to the document in the index at crawl time of a source of documents for the index;
classifying a query of the index using the metadata associated to the documents in the index, comprising;
identifying document results for the query from the index; and
identifying one or more categories for respective documents from the results using a distribution of the categories for the results provided by the metadata; and
providing one or more categories for the query based on a distribution of the one or more identified categories for respective documents in the results and
providing query results classified by one or more categories using the classification of the query.
Beschreibung
    BACKGROUND
  • [0001]
    In a computing environment, searching for particular documents and/or documents that are relevant to a particular category can involve posing a query to a search engine. Often a user may be looking for documents pertaining to a particular topic, or even for a specific document, and will create a query that helps identify the document or documents. Commonly, searches involve a set of documents, such as an Internet user searching for web-pages on a particular topic, or an employee of an enterprise searching for a document from the enterprises data storage system. In these examples, the set of documents searched by the user (web-pages on the Internet and files in a server system) are often indexed to facilitate query-based searching. An index can comprise data that represents the documents in a manner that allows query-based searching. Further, the structure of an index allows the information to be updated, such as by a web-crawler adding pages to the search index.
  • SUMMARY
  • [0002]
    This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key factors or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
  • [0003]
    Documents, such as text-based files, images, videos, and other files comprising data strings (e.g., computer program files), may be considered to belong to categories based on the topics or classes of objects that are covered in the document. Queries to a set of documents using search engines may also be considered to belong to categories based on the elements that make up the query. For example, a query may be comprised of keywords, captions, or some other data string; each of which can be considered to belong to a category individually or in combination. These categories may help characterize information or underlying intent requested by the query (e.g., “query: cardinals in ‘sports>football’” versus “query: cardinals in ‘society>religion’”), and/or an aspect of the desired results (e.g., format, reading level, authoritativeness, etc.).
  • [0004]
    Previous techniques base query classification on content of documents returned from the query. However, because classifying entire documents at query search time is computationally expensive (due to efficiency), these prior and current techniques often resort to shortcuts, such as snippet classification. The use of shortcuts in document classification typically discards information that may be relevant to a query search, and may reduce desired outcome of such query searches.
  • [0005]
    Techniques and/or systems are disclosed herein that mitigate shortcut classifications of documents returned from a query search. Further, a query classification index can be constructed that mitigates a need for performing a search engine retrieval of documents prior to classifying a query into relevant categories, for example, which can alleviate computational costs for a search query. Additionally, a number of useful applications can be layered on top a query classification distributions, which may provide more relevant search results.
  • [0006]
    In one embodiment, where documents in an index have been classified into one or more category predictions for a category hierarchy (e.g., such as classifying a web-crawl index into classes of topics), providing query results based on categorization of searched content comprises generating classification metadata for categories to which a document in the index have been classified. Further, the classification metadata can then be associated to the corresponding documents in the index, for example, by attaching the metadata to the documents. Additionally, in one embodiment, a query of the index can be classified using the metadata associated to the documents in the index, for example by category. Then query results can be provided that are classified by the one or more categories identified by the classification of the query.
  • [0007]
    To the accomplishment of the foregoing and related ends, the following description and annexed drawings set forth certain illustrative aspects and implementations. These are indicative of but a few of the various ways in which one or more aspects may be employed. Other aspects, advantages, and novel features of the disclosure will become apparent from the following detailed description when considered in conjunction with the annexed drawings.
  • DESCRIPTION OF THE DRAWINGS
  • [0008]
    FIG. 1 is a flow chart diagram of an exemplary method for providing query results based on categorization of searched content.
  • [0009]
    FIG. 2 is an exemplary embodiment of a portion of a method where query results can be provided based on classification of documents in an index.
  • [0010]
    FIG. 3 is one exemplary embodiment of a portion of a method where a query of an index can be classified using metadata associated to documents in the index.
  • [0011]
    FIG. 4 is a diagram illustrating an exemplary chart of classifications.
  • [0012]
    FIG. 5 is a flow diagram illustrating one embodiment of a portion of a method where elements of a query may be classified into categories.
  • [0013]
    FIG. 6 is a diagram of a chart illustrating an exemplary distribution of query categories.
  • [0014]
    FIG. 7 is a diagram of charts illustrating exemplary distributions of query categories.
  • [0015]
    FIG. 8 is a component block diagram of an exemplary system for providing query results based on categorization of searched content.
  • [0016]
    FIG. 9 is a component block diagram of an exemplary portion of a system, where elements of a query may be classified into categories.
  • [0017]
    FIG. 10 is an illustration of an exemplary computer-readable medium comprising processor-executable instructions configured to embody one or more of the provisions set forth herein.
  • [0018]
    FIG. 11 illustrates an exemplary computing environment wherein one or more of the provisions set forth herein may be implemented.
  • DETAILED DESCRIPTION
  • [0019]
    The claimed subject matter is now described with reference to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the claimed subject matter. It may be evident, however, that the claimed subject matter may be practiced without these specific details. In other instances, structures and devices are shown in block diagram form in order to facilitate describing the claimed subject matter.
  • [0020]
    A method may be devised that provides for constructing a query classification index that may mitigate a need to perform a search engine retrieval, which can be computationally expensive; and allows for useful applications to be layered on top of query distributions. FIG. 1 is a flow chart diagram of an exemplary method 100 for providing query results based on categorization of searched content.
  • [0021]
    The exemplary method 100 begins at 102 and involves generating classification metadata for a category to which a document in an index of documents is classified, where documents in the index have been classified into one or more category predictions for a category hierarchy. For example, web-based search engine commonly utilize a web-crawler to identify web-based documents, and the contents associated with them. The documents are often placed into a search index along with keyword content of the document identified by the web-crawler.
  • [0022]
    In one embodiment, categories may be classified by topics and the documents can be classified into a topic hierarchy, such as by general categories that contain sub-categories (e.g., sports>football>pro-football>team-name>player-name), based on the content of the document. In this embodiment, metadata that identifies a category can be created for the document, based on the one or more categories into which the document may be classified. Classification of a document can result in more than one identified category for the document, for example, where the document classification results in two or more categories meeting a desired classification threshold (e.g., having a pre-determined number of keywords associated with the category).
  • [0023]
    At 106 in the exemplary method 100, the classification metadata is associated with the document in the index. In one embodiment, the metadata may be a simple data tag that represents a classification category. In this embodiment, for respective classification categories identified for the document, the corresponding generated metadata tag can be attached to the document in the index, such as a web-search index. In this way, for example, a resulting index may comprise a list of documents, respectively having one or more (or maybe none) metadata tags attached, that can be used to identify one or more categories for the document.
  • [0024]
    In one aspect, the classification of document in an index into category predictions for a category hierarchy, such as by topics, can be undertaken in different ways. In one embodiment, a trained classifier may be used to identify one or more categories, such as by topic, for a document in the index, based on a desired threshold. For example, a web-page classifier may be trained to classify documents retrieved by a web-crawler into a hierarchy based on keywords or phrases found in the document. Typically, trained classifiers utilize a “confidence-based” classification. For example, where a confidence value can be assigned to a web-page for a particular category, the value can be a representation of how confident the classifier is that the web-page belongs to the category (e.g., 65%).
  • [0025]
    In this embodiment, a desired confidence level threshold may be set that allows a document to be classified into a particular category by the classifier. For example, where one may desire more precise search results, a ninety-five percent confidence level may be set as a threshold. Therefore, in this example, in order for a particular category to be assigned to a document the classifier needs to be at least ninety-five percent confident that the document belongs to a category (e.g., based on the keywords and/or phrases found in the document that correlate to a category, such as “football” is associated with “sports”).
  • [0026]
    Further, in this aspect, in one embodiment, a retrieval relevance score can be used to classify categories for a document. A score can be determined for a document to identify one or more categories for the document, based on a desired threshold. Typical relevance scores comprise variables such as a document's on-page factors, keywords, content, and if online, the page's link popularity, user engagement and competition for related phrases. As an example, these variables can be assigned a value and run through an algorithm that determines a relevance score for a category for a document. In this embodiment, if the relevance score for a category, such as a topic, meets a desired threshold, the category may be assigned to the document.
  • [0027]
    It will be appreciated that the methods and systems, described herein, are not limited to the embodiments of indexed documents described above. It is anticipated that those skilled in the art may devise alternate document indices that can utilize the techniques described herein. For example, document classification can be extended to enterprise data searching, desktop (or laptop) searching, searching over a private collection, searching of a federated collection of document, and more. Further, the index may or may not be generated from an online crawl, and the searching may or may not be conducted online.
  • [0028]
    In another aspect, a source collection of documents that is subject to a classification may be comprised of a variety of collections that are federated to an index. For example, a knowledge-based search, such as a library, may utilize and index that comprises documents from internal archives, online documents, other library collections, private collections, and more. In one embodiment, while the collections of documents may be disparate, and may comprise a variety of potential categories, they can be federated together and classified to one or more categories, as described above. Further, in another embodiment, the separate collections may be indexed and custom categories can be created for the respective collections.
  • [0029]
    In another aspect, category classification and metadata association for respective documents in an index may be performed at crawl-time of the document source (e.g., the collection of documents crawled to create/update the index), or at a desired time after the index is created. In one embodiment, respective documents may be classified into categories as they are identified by a crawler. In this embodiment, upon classification at crawl-time, the metadata can be generated for the respective classification categories, and the metadata can be associated to the document in the index. In this way, for example, the index for the documents is created at crawl-time, along with the associated metadata that described the categories for the respective documents.
  • [0030]
    In FIG. 1, having associated metadata with classified documents in an index, the exemplary method 100 ends at 108.
  • [0031]
    In one aspect, an index of documents may be utilized in a search query for relevant documents. An index of documents may represent an entirety of the documents available for a search, or can represent a sub-set of the documents that one may wish to query. For example, an index that is used by an online search engine merely comprises those web-pages crawled by its web-crawler from all of websites connected to the Internet. However, an index of an enterprise data base may comprise representations of the respective documents stored for the enterprise.
  • [0032]
    In this aspect, in one embodiment, one may use the metadata associated with the documents in an index to classify a query and predict appropriate query results. FIG. 2 is an exemplary embodiment of portion of a method 200 where query results can be provided based on the classification of documents in an index. In the exemplary embodiment, a query is issued 250, for example, by a user wishing to identify relevant documents from a set of documents (e.g., a user wishing to finding web-pages showing college football scores on the Internet; or a CFO wishing to find monthly financial reports for particular offices on an enterprise data management and storage system).
  • [0033]
    In this embodiment 200, a search engine can use the query 250 to search an index, and the query can be classified using the metadata associated to the documents in the index, at 202. For example, a one word query, such as “bass,” may be classified into a plurality of topic categories such as fish>bass, stringed-instrument>bass, and men's shoes>bass. Further, the respective topic categories may be sub-topics in one or more larger categories, such as outdoor recreation>sports>fishing>fresh water>fish>bass, arts>music>musical instruments>stringed instruments>bass, and shopping>clothing>footwear>shoes>men's shoes>bass.
  • [0034]
    In this embodiment, respective keywords or combinations of keywords in a query can be used to identify documents in the index, based on the content keywords and phrases stored by the index for the respective pages. Further, the metadata that represents the categories associated with the documents can be used to classify the respective query keywords and/or combinations, for example, by counting a how many times a category is identified as associated with a document called by query.
  • [0035]
    FIG. 3 is one exemplary embodiment 300 of a portion of a method where a query of the index can be classified using the metadata associated to the documents in the index. The when a query 350 is posed, document results 352 for the query can be identified from the index, at 302, such as by using a search engine. At 304, one or more categories for the respective documents results can be identified using a distribution of the categories that is provided by the metadata attached to respective documents.
  • [0036]
    FIG. 4 is a diagram illustrating an exemplary chart 400 of category classification. In this exemplary chart 400, a query “bass” may be posed to a web-based search engine. Classification category topics 402 can come from metadata attached to documents in the index, and the topics corresponding to documents identified as having content related to the query are shown in a distribution 404. Therefore, in this example, the query has been classified into the topics 402 identified by using the metadata associated to the documents in the index.
  • [0037]
    Returning to FIG. 3, at 306, one or more categories 354 for the query can be provided (e.g., predicted) based on a distribution of the one or more identified categories for respective documents in the results. For example, as illustrated in FIG. 4, the topics may be distributed by rank 404 based on how many documents returned by a query are classified into respective topics for the query. In the exemplary chart 400, a percentage 406 of a total number of documents returned by the query for the topics can provide a distribution ranking 404 of the respective topics 402. In this example, “recreation/outdoors” is attached to thirty percent of the documents returned from the index, “arts” is attached to twenty-five percent of the documents returned from the index, and so-on.
  • [0038]
    In one embodiment, the number of provided query categories 354 may result from a cut-off threshold value, where a desired percentage/ranking for the categories is needed for the document to be included in the returned results. For example, in FIG. 4, a query of “bass” may return the categories 354 from merely the topics 402 recreation/outdoors, arts, and recreation.
  • [0039]
    Returning to FIG. 2, at 204, query results can be provided that are classified by one or more categories using the classification of the query. In one embodiment, the query results provided (e.g., the documents from the index) may be representative of categories that have a desired number (e.g., top percentage) of documents associated with the category. In another embodiment, a search engine may return results for the respective categories, and may also return merely categories that meet a desired ranking/percentage threshold. In another example, the search engine may return results for the respective categories.
  • [0040]
    As described above, a set of documents that is represented by an index may comprise one or more types of documents from one or more collections. For example, documents can comprise text files, spreadsheets, presentations, web-pages, photos, images, videos, programs, and many more. Further, documents can be found on the Internet as part of an advertisement, social network, virtual storage, websites, video sites, for example. Therefore, when a query is performed for documents the query can be posed in different ways and may be comprised of a variety of elements.
  • [0041]
    In one embodiment, a query may be comprised of keyword terms (and/or combinations of terms) for a search-based query, such as for an Internet-based search engine. A query can also be comprised of keywords associated with Internet-based advertisements, such as when an advertisement is placed on a search result page based on a keyword. Further, a query can be comprised of tag terms associated with a file, such as when searching an enterprise database for a word file containing particular terms. Additionally, a query can be comprised of keywords associated with a caption, such as for a photo or video.
  • [0042]
    It will be appreciated that the techniques and systems described herein are not limited to these embodiments of document types and query elements. Those skilled in the art may devise alternate document types and query elements that can utilize the techniques described above. The embodiments described above are merely examples of how the techniques may be applied.
  • [0043]
    In one aspect, a query classification index can be created that maps words and phrases to category distributions, which may allow for efficient query classification without having to perform a search engine retrieval, for example. FIG. 5 is a flow diagram illustrating one embodiment of a portion of a method 500 where elements of a query may be classified into categories. The exemplary embodiment of the method 500 begins at 502 and involves classifying documents that are in an index into categories. As described above, a set of documents (e.g., web-pages, enterprise documents) can be stored as an index and classified into topics, for example, by a using a trained classifier and/or by determining a retrieval relevance score. For example, a naive-Bayes or a logistic regression classifier can be trained over a crawl of a set of documents and used to label the respective documents in the index with one to three category predictions, depending on a desired threshold.
  • [0044]
    After classification, a mapping of query terms to a category distribution for the index can be built. At 506, terms (e.g., words, phrases, captions, data-strings, etc.) found in a document (e.g., text document, image, video, other file) can be mapped to the one or more category predictions for the document. In one embodiment, the respective documents in the index can be scanned for terms (e.g., keywords), for example, by using a unigram language model to evaluate respective terms independently of context with other terms in the document. In this embodiment, the terms identified in a document can be mapped to the one or more category predictions associated with the document in the index.
  • [0045]
    At 508, the terms mapped for the index can be aggregated to create a distribution of category predictions for the respective terms, and a category prediction distribution can be created for the aggregated terms, at 510. For example, a term “bass” may be identified in a plurality of documents in the index. In this embodiment, the respective category predictions for the documents in which “bass” is found can be aggregated to create a topic distribution for “bass”, such as shown in the exemplary chart 400, in FIG. 4. In this way, in this example, a topic distribution for the term “bass” can be created without performing a query search for the term.
  • [0046]
    When building a mapping of terms, a query of the index can be classified using the metadata associated to the documents in the index. At 512, when a query is posed, respective query terms (and/or combinations of terms) can be mapped to the distribution of category predictions for the respective terms in the index. For example, if “bass” is a term in a posed query, it can be mapped to the category predictions distribution created for “bass” (e.g., as in FIG. 4, 400). Further, if a query comprises more than one term, the respective terms can be mapped to corresponding category predictions distributions.
  • [0047]
    At 514, one or more categories for the query can be identified using the query term mapping. In one embodiment, respective terms in a query (one or more) can each elicit category predictions. For example, a query for “apple” may result in topics such as food, agriculture, computers, electronics, etc. Further, a query for “apple pie” may elicit those for “apple” plus additional topics for “pie,” such as home, shopping, cooking, dessert, etc.
  • [0048]
    At 516, one or more predicted relevant topic models for the query can be provided based on the distribution of the identified categories for the query. In one embodiment, topic models may comprise category predictions from an intersection of the sets of categories for a multi-term query. For example, both “apple” and “pie” category predictions may include “food” and “cooking,” where as category predictions for “apple iPod” may both include “computers” & “electronics.”
  • [0049]
    In another embodiment, a desired number (e.g., top 10) of category predictions can be selected for relevant topic models. For example, those categories that meet a threshold of percentage of all topics (e.g., as in FIG. 4, 406) may be selected for the respective terms in the query. In another embodiment, a combination of intersecting topics and those meeting a threshold can be selected. In this way, for example, when a query is posed, relevant topic models can be predicted prior to a search of the index, which can increase efficiency of the search of the index.
  • [0050]
    In one aspect, query category distributions may be utilized to enhance query output from a search of a set of documents. In one embodiment, query ambiguity may be determined for a query by using an ambiguity function that accounts for a spread of a distribution of categories for the query, for example, to generate an ambiguity score for the query. For example, after a query classified into topic predictions, one may use a spread of the distributions of the predicted topics to evaluate entropy, or a similar measure, between the topics in the distribution.
  • [0051]
    FIG. 6 is a diagram of a chart illustrating an exemplary distribution 600 of query topics. In this example, the distribution 600 can represent a query for large a computer technology company. Resulting category predictions 602 show expected topics for such a query, along with the graphic ranking 604 of the topics, and the percentage of documents 606 from the set of documents in the index that are associated with the respective categories. The exemplary distribution 600 illustrates that seventy-three percent of the documents are associate with “computers” with the remaining documents being associated with the other topics at much smaller percentages.
  • [0052]
    In this example, a resulting ambiguity score may be determined by applying an ambiguity function to a desired number of categories in the distribution. For example, the function may account for a top three categories (e.g., computers, business, and society in FIG. 6) to determine if the query is ambiguous. In FIG. 6, the function may predict that the query for the large computer technology company is not ambiguous due to the distribution. However, applying an ambiguity function to the query “bass” as in FIG. 4, 400, may yield a result, such as a score, that is more ambiguous than the distribution in 600.
  • [0053]
    In another embodiment, in this aspect, a query similarity between the query and an alternate query may be determined by comparing the distribution of the categories associated with the query and the distribution of the categories associated with the alternate query. For example, a first query for “bad credit mortgage” may result in a similar topic distribution as a second query for “refinance.” FIG. 7 is a diagram of charts 700 and 750 illustrating exemplary distributions of query topics.
  • [0054]
    In FIG. 7, the distribution for the query “bad credit mortgage” can be shown in the chart 700, and the distribution for the query “refinance” can be shown in the chart 750. In the exemplary distributions of topics 702 and 752, similar topics have been identified for the queries. Further, the rankings 704 and 754, and the percentage of documents for respective topics 706 and 756 have similar distributions for the top categories “business/financial service” and “business.”
  • [0055]
    In this embodiment, a query similarity score for these two queries, for example, can be determined by calculating a similarity function using the distribution of the categories associated with the first query and the distribution of the categories associated with the second query. In this way, in this example, a similarity score for “bad credit mortgage” and “refinance” may show more similarity that one between the queries in FIGS. 4 and 6.
  • [0056]
    In another embodiment, the determination of query ambiguity for a query can be combined with the determination of query similarity between the query and an alternate query to yield a query quality characteristic. In this embodiment, query quality may be predicted in a variety of ways. For example, where a query has is very precise (e.g., low ambiguity score) but does not have a good match with an alternate query (e.g., high query similarity score) one can have a higher confidence that the two queries do not match. In this example, the first query is not ambiguous so it has a low chance of having multiple meanings, and therefore is less likely to have mistaken category predictions for the similarity prediction.
  • [0057]
    On the other hand, if the first query is ambiguous and is a bad match, the poor similarity prediction may be due to the query's ambiguity. In one embodiment, an ambiguity score and similarity score may be combined resulting in an overall quality score for a query, when comparing it to another query. For example, an online advertiser may want to mitigate a chance that their word-based ads are not erroneously associated with a search engine query that is not topical to their advertisement. In this embodiment, an overall quality score can be generated for the ad as compared with the search engine query.
  • [0058]
    In another aspect, a query refinement score can be determined for an alternate query (e.g., a refinement of a first query that is composed after the first query does not return desired results) that can be used to measure whether the alternate query is a more specific refinement of one or more portions of the query, a more general refinement of one or more portions of the query, or an orthogonal concept to one or more portions of the query (e.g., neither more general nor more specific). For example, where a first query may be for “football,” and alternate query for “professional football” would be a more specific refinement, an alternate query for “sports” would be a more general refinement, and an alternate query for “telephones” would be orthogonal. In this way, a query refinement score may facilitate query classification by utilizing existing topic distributions for the query, or having to begin a new query classification.
  • [0059]
    A system may be devised that helps characterize intent of a query of a set of documents by constructing a query classification index for example, and mapping query terms to the constructed index for classification into categories. FIG. 8 is a component block diagram of an exemplary system 800 for providing query results based on categorization of searched content, such as an index of documents. For example, a set of documents (e.g., web-pages, enterprise files, image and video libraries) may be crawled and indexed for later retrieval using a search engine.
  • [0060]
    The exemplary system 800 comprises computer-based storage 802 that is configured to store classification metadata associated with documents in an index of documents 850. Further, where documents in the index 850 have been classified into one or more category predictions for a category hierarchy, a metadata generator 804 is configured to generate classification metadata for a document in the index, which identifies a category to which the document in the index 850 has been classified. For example, a trained classifier 812 may be used to classify documents, from a set of documents, into one or more categories, and the metadata generator 804 can generate metadata that identifies the one or categories.
  • [0061]
    Additionally, a metadata association component 806, which is operably coupled to the storage 802, is configured to associate the classification metadata to the corresponding document in the index 850. Therefore, for example, the index 850 may comprise documents that have attached metadata identifying their respective classification topics.
  • [0062]
    In one embodiment, the exemplary system 800 comprises a query classifier 808 that is configured to classify a query 852 of the index 850 using the metadata associated to the documents in the index, in association with results of the query 854. For example, where a query may return resulting documents based on the search engine used, the metadata attached to the returned documents can be used by the query classifier 808 to classify the query into categories. Further, in this embodiment, the query classifier 808 can comprise a query distribution component that is configured to generate a distribution of categories (e.g., a ranking of topics based on an aggregation of metadata for the returned documents), provided by the metadata, for query results.
  • [0063]
    Additionally, in this embodiment, a classified query results provider 810, which is operably coupled to the query classifier 808, can provide query results classified by one or more categories 856 using the classification of the query from the query classifier 808. For example, the query results provider 810 may be set to provide query results for a desired number or categories that meet a threshold for number of documents in a category (e.g., top three categories by percentage of documents in the category, such as in chart 400 of FIG. 4). In this way, in this example, merely those categories that comprise a desired number of matching documents will have results returned in the classified results 856.
  • [0064]
    FIG. 9 is a component block diagram of an exemplary portion of a system 900, where elements of a query may be classified into categories. In the exemplary portion of a system 900, a document term mapper 902 is configured to map respective terms in a document in the index 850 to the one or more category predictions for the document. Further, the document term mapper 902 is configured to aggregate the terms mapped for the index 850 to create a distribution of category predictions for the respective terms 952. In one embodiment, the category predictions for the respective terms 952 can be stored in the data storage component 802 for the system.
  • [0065]
    Additionally, in the exemplary portion of the system 900, a mapped query classifier 904 configured to map respective query terms (e.g., words, data strings, captions, etc.), for a query 954 of the index 850, to the distribution of category predictions, such as those in storage 802 from the document term mapper 902, for the respective terms in the index 850. The mapped query classifier 904 is also configured to identify one or more categories for the query 954 using the query term mapping, and provide one or more predicted relevant topic models 956 for the query based on the distribution of the identified categories for the query.
  • [0066]
    Still another embodiment involves a computer-readable medium comprising processor-executable instructions configured to implement one or more of the techniques presented herein. An exemplary computer-readable medium that may be devised in these ways is illustrated in FIG. 10, wherein the implementation 1000 comprises a computer-readable medium 1008 (e.g., a CD-R, DVD-R, or a platter of a hard disk drive), on which is encoded computer-readable data 1006. This computer-readable data 1006 in turn comprises a set of computer instructions 1004 configured to operate according to one or more of the principles set forth herein. In one such embodiment 1002, the processor-executable instructions 1004 may be configured to perform a method, such as the exemplary method 100 of FIG. 1, for example. In another such embodiment, the processor-executable instructions 1004 may be configured to implement a system, such as the exemplary system 800 of FIG. 8, for example. Many such computer-readable media may be devised by those of ordinary skill in the art that are configured to operate in accordance with the techniques presented herein.
  • [0067]
    Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
  • [0068]
    As used in this application, the terms “component,” “module,” “system”, “interface”, and the like are generally intended to refer to a computer-related entity, either hardware, a combination of hardware and software, software, or software in execution. For example, a component may be, but is not limited to being, a process running on a processor, a processor, an object, an executable, a thread of execution, a program, and/or a computer. By way of illustration, both an application running on a controller and the controller can be a component. One or more components may reside within a process and/or thread of execution and a component may be localized on one computer and/or distributed between two or more computers.
  • [0069]
    Furthermore, the claimed subject matter may be implemented as a method, apparatus, or article of manufacture using standard programming and/or engineering techniques to produce software, firmware, hardware, or any combination thereof to control a computer to implement the disclosed subject matter. The term “article of manufacture” as used herein is intended to encompass a computer program accessible from any computer-readable device, carrier, or media. Of course, those skilled in the art will recognize many modifications may be made to this configuration without departing from the scope or spirit of the claimed subject matter.
  • [0070]
    FIG. 11 and the following discussion provide a brief, general description of a suitable computing environment to implement embodiments of one or more of the provisions set forth herein. The operating environment of FIG. 11 is only one example of a suitable operating environment and is not intended to suggest any limitation as to the scope of use or functionality of the operating environment. Example computing devices include, but are not limited to, personal computers, server computers, hand-held or laptop devices, mobile devices (such as mobile phones, Personal Digital Assistants (PDAs), media players, and the like), multiprocessor systems, consumer electronics, mini computers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
  • [0071]
    Although not required, embodiments are described in the general context of “computer readable instructions” being executed by one or more computing devices. Computer readable instructions may be distributed via computer readable media (discussed below). Computer readable instructions may be implemented as program modules, such as functions, objects, Application Programming Interfaces (APIs), data structures, and the like, that perform particular tasks or implement particular abstract data types. Typically, the functionality of the computer readable instructions may be combined or distributed as desired in various environments.
  • [0072]
    FIG. 11 illustrates an example of a system 1110 comprising a computing device 1112 configured to implement one or more embodiments provided herein. In one configuration, computing device 1112 includes at least one processing unit 1116 and memory 1118. Depending on the exact configuration and type of computing device, memory 1118 may be volatile (such as RAM, for example), non-volatile (such as ROM, flash memory, etc., for example) or some combination of the two. This configuration is illustrated in FIG. 11 by dashed line 1114.
  • [0073]
    In other embodiments, device 1112 may include additional features and/or functionality. For example, device 1112 may also include additional storage (e.g., removable and/or non-removable) including, but not limited to, magnetic storage, optical storage, and the like. Such additional storage is illustrated in FIG. 11 by storage 1120. In one embodiment, computer readable instructions to implement one or more embodiments provided herein may be in storage 1120. Storage 1120 may also store other computer readable instructions to implement an operating system, an application program, and the like. Computer readable instructions may be loaded in memory 1118 for execution by processing unit 1116, for example.
  • [0074]
    The term “computer readable media” as used herein includes computer storage media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions or other data. Memory 1118 and storage 1120 are examples of computer storage media. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVDs) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by device 1112. Any such computer storage media may be part of device 1112.
  • [0075]
    Device 1112 may also include communication connection(s) 1126 that allows device 1112 to communicate with other devices. Communication connection(s) 1126 may include, but is not limited to, a modem, a Network Interface Card (NIC), an integrated network interface, a radio frequency transmitter/receiver, an infrared port, a USB connection, or other interfaces for connecting computing device 1112 to other computing devices. Communication connection(s) 1126 may include a wired connection or a wireless connection. Communication connection(s) 1126 may transmit and/or receive communication media.
  • [0076]
    The term “computer readable media” may include communication media. Communication media typically embodies computer readable instructions or other data in a “modulated data signal” such as a carrier wave or other transport mechanism and includes any information delivery media. The term “modulated data signal” may include a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal.
  • [0077]
    Device 1112 may include input device(s) 1124 such as keyboard, mouse, pen, voice input device, touch input device, infrared cameras, video input devices, and/or any other input device. Output device(s) 1122 such as one or more displays, speakers, printers, and/or any other output device may also be included in device 1112. Input device(s) 1124 and output device(s) 1122 may be connected to device 1112 via a wired connection, wireless connection, or any combination thereof. In one embodiment, an input device or an output device from another computing device may be used as input device(s) 1124 or output device(s) 1122 for computing device 1112.
  • [0078]
    Components of computing device 1112 may be connected by various interconnects, such as a bus. Such interconnects may include a Peripheral Component Interconnect (PCI), such as PCI Express, a Universal Serial Bus (USB), firewire (IEEE 1394), an optical bus structure, and the like. In another embodiment, components of computing device 1112 may be interconnected by a network. For example, memory 1118 may be comprised of multiple physical memory units located in different physical locations interconnected by a network.
  • [0079]
    Those skilled in the art will realize that storage devices utilized to store computer readable instructions may be distributed across a network. For example, a computing device 1130 accessible via network 1128 may store computer readable instructions to implement one or more embodiments provided herein. Computing device 1112 may access computing device 1130 and download a part or all of the computer readable instructions for execution. Alternatively, computing device 1112 may download pieces of the computer readable instructions, as needed, or some instructions may be executed at computing device 1112 and some at computing device 1130.
  • [0080]
    Various operations of embodiments are provided herein. In one embodiment, one or more of the operations described may constitute computer readable instructions stored on one or more computer readable media, which if executed by a computing device, will cause the computing device to perform the operations described. The order in which some or all of the operations are described should not be construed as to imply that these operations are necessarily order dependent. Alternative ordering will be appreciated by one skilled in the art having the benefit of this description. Further, it will be understood that not all operations are necessarily present in each embodiment provided herein.
  • [0081]
    Moreover, the word “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous over other aspects or designs. Rather, use of the word exemplary is intended to present concepts in a concrete fashion. As used in this application, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or”. That is, unless specified otherwise, or clear from context, “X employs A or B” is intended to mean any of the natural inclusive permutations. That is, if X employs A; X employs B; or X employs both A and B, then “X employs A or B” is satisfied under any of the foregoing instances. In addition, the articles “a” and “an” as used in this application and the appended claims may generally be construed to mean “one or more” unless specified otherwise or clear from context to be directed to a singular form.
  • [0082]
    Also, although the disclosure has been shown and described with respect to one or more implementations, equivalent alterations and modifications will occur to others skilled in the art based upon a reading and understanding of this specification and the annexed drawings. The disclosure includes all such modifications and alterations and is limited only by the scope of the following claims. In particular regard to the various functions performed by the above described components (e.g., elements, resources, etc.), the terms used to describe such components are intended to correspond, unless otherwise indicated, to any component which performs the specified function of the described component (e.g., that is functionally equivalent), even though not structurally equivalent to the disclosed structure which performs the function in the herein illustrated exemplary implementations of the disclosure. In addition, while a particular feature of the disclosure may have been disclosed with respect to only one of several implementations, such feature may be combined with one or more other features of the other implementations as may be desired and advantageous for any given or particular application. Furthermore, to the extent that the terms “includes”, “having”, “has”, “with”, or variants thereof are used in either the detailed description or the claims, such terms are intended to be inclusive in a manner similar to the term “comprising.”
Patentzitate
Zitiertes PatentEingetragen Veröffentlichungsdatum Antragsteller Titel
US5926808 *25. Juli 199720. Juli 1999Claritech CorporationDisplaying portions of text from multiple documents over multiple databases related to a search query in a computer network
US6321224 *28. Febr. 200020. Nov. 2001Requisite Technology, Inc.Database search, retrieval, and classification with sequentially applied search algorithms
US6631496 *22. März 19997. Okt. 2003Nec CorporationSystem for personalizing, organizing and managing web information
US7496567 *28. Sept. 200524. Febr. 2009Terril John SteichenSystem and method for document categorization
US7574433 *8. Okt. 200411. Aug. 2009Paterra, Inc.Classification-expanded indexing and retrieval of classified documents
US7620628 *6. Dez. 200417. Nov. 2009Yahoo! Inc.Search processing with automatic categorization of queries
US7769760 *10. Sept. 20073. Aug. 2010Sony CorporationInformation processing apparatus, method and program thereof
US7774341 *6. März 200710. Aug. 2010Veveo, Inc.Methods and systems for selecting and presenting content based on dynamically identifying microgenres associated with the content
US7779009 *27. Jan. 200617. Aug. 2010Aol Inc.Web query classification
US7801896 *19. Febr. 200721. Sept. 2010Andrew J SzaboDatabase access system
US8010545 *15. Jan. 200930. Aug. 2011Palo Alto Research Center IncorporatedSystem and method for providing a topic-directed search
US8041733 *9. Febr. 200918. Okt. 2011Yahoo! Inc.System for automatically categorizing queries
US20030061243 *26. Aug. 200227. März 2003Kim Jeong JungInformation auto classification method and information search and analysis method
US20030172357 *11. März 200211. Sept. 2003Kao Anne S.W.Knowledge management using text classification
US20050055341 *22. Sept. 200310. März 2005Paul HaahrSystem and method for providing search query refinements
US20070038601 *10. Aug. 200515. Febr. 2007Guha Ramanathan VAggregating context data for programmable search engines
US20070156677 *19. Febr. 20075. Juli 2007Alberti Anemometer LlcDatabase access system
US20080065659 *10. Sept. 200713. März 2008Akihiro WatanabeInformation processing apparatus, method and program thereof
US20080201219 *20. Febr. 200721. Aug. 2008Andrei Zary BroderQuery classification and selection of associated advertising information
US20080215557 *30. Okt. 20074. Sept. 2008Jorey RamerMethods and systems of mobile query classification
US20080270462 *6. Nov. 200730. Okt. 2008Interse A/SSystem and Method of Uniformly Classifying Information Objects with Metadata Across Heterogeneous Data Stores
US20090094223 *5. Okt. 20079. Apr. 2009Matthew BerkSystem and method for classifying search queries
Referenziert von
Zitiert von PatentEingetragen Veröffentlichungsdatum Antragsteller Titel
US834111219. Mai 200625. Dez. 2012Microsoft CorporationAnnotation by search
US8396878 *26. Sept. 201112. März 2013Limelight Networks, Inc.Methods and systems for generating automated tags for video files
US84429517. Dez. 201114. Mai 2013International Business Machines CorporationProcessing archive content based on hierarchical classification levels
US848401514. Mai 20109. Juli 2013Wolfram Alpha LlcEntity pages
US849502727. Juni 201223. Juli 2013International Business Machines CorporationProcessing archive content based on hierarchical classification levels
US85596829. Nov. 201015. Okt. 2013Microsoft CorporationBuilding a person profile database
US85898697. Sept. 200719. Nov. 2013Wolfram Alpha LlcMethods and systems for determining a formula
US8601015 *14. Mai 20103. Dez. 2013Wolfram Alpha LlcDynamic example generation for queries
US8707192 *21. Okt. 201022. Apr. 2014Apple Inc.Browsing or searching user interfaces and other aspects
US871346213. Okt. 201029. Apr. 2014Apple Inc.Browsing or searching user interfaces and other aspects
US873260027. Okt. 201020. Mai 2014Apple Inc.Browsing or searching user interfaces and other aspects
US8788514 *28. Okt. 201022. Juli 2014Google Inc.Triggering music answer boxes relevant to user search queries
US879930622. Dez. 20115. Aug. 2014Alibaba Group Holding LimitedRecommendation of search keywords based on indication of user intention
US8799799 *13. Juni 20135. Aug. 2014Palantir Technologies Inc.Interactive geospatial map
US881229828. Juli 201019. Aug. 2014Wolfram Alpha LlcMacro replacement of natural language input
US88129603. Dez. 201319. Aug. 2014Palantir Technologies Inc.Cohort-based presentation of user interaction data
US883259420. Dez. 20139. Sept. 2014Palantir Technologies Inc.Space-optimized display of multi-column tables with selective text truncation based on a combined text width
US8838597 *16. Aug. 201116. Sept. 2014Semantifi, Inc.Semantic metadata extraction
US8843470 *5. Okt. 201223. Sept. 2014Microsoft CorporationMeta classifier for query intent classification
US88559995. Febr. 20147. Okt. 2014Palantir Technologies Inc.Method and system for generating a parser and parsing complex data
US886848610. Dez. 201321. Okt. 2014Palantir Technologies Inc.Time-sensitive cube
US889818011. Jan. 201025. Nov. 2014Alibaba Group Holding LimitedMethod and system for querying information
US890379828. Mai 20102. Dez. 2014Microsoft CorporationReal-time annotation and enrichment of captured video
US891727419. Dez. 201323. Dez. 2014Palantir Technologies Inc.Event matrix based on integrated data
US892487216. Dez. 201330. Dez. 2014Palantir Technologies Inc.Overview user interface of emergency call data of a law enforcement agency
US89308972. Okt. 20136. Jan. 2015Palantir Technologies Inc.Data integration tool
US89376196. Jan. 201420. Jan. 2015Palantir Technologies Inc.Generating an object time series from data objects
US893847527. Dez. 201120. Jan. 2015Sap SeManaging business objects data sources
US89386863. Okt. 201320. Jan. 2015Palantir Technologies Inc.Systems and methods for analyzing performance of an entity
US896638921. Sept. 200724. Febr. 2015Limelight Networks, Inc.Visual interface for identifying positions of interest within a sequentially ordered information encoding
US896643918. Nov. 201324. Febr. 2015Wolfram Alpha LlcMethod and system for determining an answer to a query
US90091712. Mai 201414. Apr. 2015Palantir Technologies Inc.Systems and methods for active column filtering
US900982716. Mai 201414. Apr. 2015Palantir Technologies Inc.Security sharing system
US9015152 *20. Juli 201121. Apr. 2015Google Inc.Managing search results
US901517215. Juni 201221. Apr. 2015Limelight Networks, Inc.Method and subsystem for searching media content within a content-search service system
US9020936 *14. Aug. 200928. Apr. 2015Microsoft Technology Licensing, LlcUsing categorical metadata to rank search results
US902126029. Aug. 201428. Apr. 2015Palantir Technologies Inc.Malware data item analysis
US90213842. Juni 201428. Apr. 2015Palantir Technologies Inc.Interactive vehicle information map
US904369627. Febr. 201426. Mai 2015Palantir Technologies Inc.Systems and methods for visual definition of data associations
US90438946. Febr. 201526. Mai 2015Palantir Technologies Inc.Malicious software detection in a computing system
US906981427. Juli 201230. Juni 2015Wolfram Alpha LlcMethod and system for using natural language to generate widgets
US9092478 *27. Dez. 201128. Juli 2015Sap SeManaging business objects data sources
US91169751. Okt. 201425. Aug. 2015Palantir Technologies Inc.Systems and user interfaces for dynamic and interactive simultaneous querying of multiple data stores
US9116996 *24. Juli 201225. Aug. 2015Google Inc.Reverse question answering
US912308615. März 20131. Sept. 2015Palantir Technologies, Inc.Automatically generating event objects from images
US912921930. Juni 20148. Sept. 2015Palantir Technologies, Inc.Crime risk forecasting
US919574124. Juni 201424. Nov. 2015Google Inc.Triggering music answer boxes relevant to user search queries
US920224929. Aug. 20141. Dez. 2015Palantir Technologies Inc.Data item clustering and analysis
US921376814. Mai 201015. Dez. 2015Wolfram Alpha LlcAssumption mechanism for queries
US92237736. Jan. 201429. Dez. 2015Palatir Technologies Inc.Template system for custom document generation
US923563812. Nov. 201312. Jan. 2016International Business Machines CorporationDocument retrieval using internal dictionary-hierarchies to adjust per-subject match results
US92398486. Febr. 201219. Jan. 2016Microsoft Technology Licensing, LlcSystem and method for semantically annotating images
US925113616. Okt. 20132. Febr. 2016International Business Machines CorporationDocument tagging and retrieval using entity specifiers
US92566643. Juli 20149. Febr. 2016Palantir Technologies Inc.System and method for news events detection and visualization
US926251010. Mai 201316. Febr. 2016International Business Machines CorporationDocument tagging and retrieval using per-subject dictionaries including subject-determining-power scores for entries
US92986789. Juli 201429. März 2016Palantir Technologies Inc.System and method for news events detection and visualization
US933589720. Sept. 201310. Mai 2016Palantir Technologies Inc.Long click display of a context menu
US9335911 *18. Sept. 201510. Mai 2016Palantir Technologies Inc.Interactive user interface for dynamic data analysis exploration and query processing
US934444715. Sept. 201417. Mai 2016Palantir Technologies Inc.Internal malware data item clustering and analysis
US936787222. Dez. 201414. Juni 2016Palantir Technologies Inc.Systems and user interfaces for dynamic and interactive investigation of bad actor behavior based on automatic clustering of related data in various data structures
US938043115. März 201328. Juni 2016Palantir Technologies, Inc.Use of teams in a mobile application
US93839119. Sept. 20095. Juli 2016Palantir Technologies, Inc.Modal-less interface enhancements
US940542429. Aug. 20122. Aug. 2016Wolfram Alpha, LlcMethod and system for distributing and displaying graphical items
US9411905 *26. Sept. 20139. Aug. 2016Groupon, Inc.Multi-term query subsumption for document classification
US943055915. Sept. 201530. Aug. 2016International Business Machines CorporationDocument retrieval using internal dictionary-hierarchies to adjust per-subject match results
US943056816. Sept. 201430. Aug. 2016Alibaba Group Holding LimitedMethod and system for querying information
US944903511. März 201520. Sept. 2016Palantir Technologies Inc.Systems and methods for active column filtering
US945428131. Aug. 201527. Sept. 2016Palantir Technologies Inc.System for providing dynamic linked panels in user interface
US945478517. Sept. 201527. Sept. 2016Palantir Technologies Inc.Systems and user interfaces for holistic, data-driven investigation of bad actor behavior based on clustering and scoring of related data
US94601753. Juni 20154. Okt. 2016Palantir Technologies Inc.Server implemented geographic information system with graphical interface
US948316230. Juni 20141. Nov. 2016Palantir Technologies Inc.Relationship visualizations
US948354615. Dez. 20141. Nov. 2016Palantir Technologies Inc.System and method for associating related records to common entities across multiple lists
US950185130. Sept. 201522. Nov. 2016Palantir Technologies Inc.Time-series analysis system
US951420031. Juli 20156. Dez. 2016Palantir Technologies Inc.Systems and user interfaces for dynamic and interactive simultaneous querying of multiple data stores
US95144141. Apr. 20166. Dez. 2016Palantir Technologies Inc.Systems and methods for identifying and categorizing electronic documents through machine learning
US95526155. März 201524. Jan. 2017Palantir Technologies Inc.Automated database analysis to detect malfeasance
US95578821. Apr. 201431. Jan. 2017Palantir Technologies Inc.Context-sensitive views
US955835228. Apr. 201531. Jan. 2017Palantir Technologies Inc.Malicious software detection in a computing system
US9576052 *16. Juli 201321. Febr. 2017Xerox CorporationSystems and methods of web crawling
US958929911. Mai 20167. März 2017Palantir Technologies Inc.Systems and user interfaces for dynamic and interactive investigation of bad actor behavior based on automatic clustering of related data in various data structures
US96001465. Nov. 201521. März 2017Palantir Technologies Inc.Interactive geospatial map
US961955730. Juni 201411. Apr. 2017Palantir Technologies, Inc.Systems and methods for key phrase characterization of documents
US96395802. Nov. 20152. Mai 2017Palantir Technologies, Inc.Computer-implemented systems and methods for data management and visualization
US964639614. Jan. 20159. Mai 2017Palantir Technologies Inc.Generating object time series and data objects
US965244419. Nov. 201416. Mai 2017Microsoft Technology Licensing, LlcReal-time annotation and enrichment of captured video
US9652527 *30. Juni 201616. Mai 2017Groupon, Inc.Multi-term query subsumption for document classification
US9652547 *15. Sept. 201416. Mai 2017Google Inc.Framework for suggesting search terms
US967899218. Mai 201113. Juni 2017Microsoft Technology Licensing, LlcText to image translation
US968472123. Febr. 201520. Juni 2017Wolfram Alpha LlcPerforming machine actions in response to voice input
US970378228. Mai 201011. Juli 2017Microsoft Technology Licensing, LlcAssociating media with metadata of near-duplicates
US972756025. Febr. 20158. Aug. 2017Palantir Technologies Inc.Systems and methods for organizing and identifying documents via hierarchies and dimensions of tags
US972762216. Juni 20148. Aug. 2017Palantir Technologies, Inc.Methods and systems for analyzing entity performance
US973421716. Juni 201415. Aug. 2017Palantir Technologies Inc.Methods and systems for analyzing entity performance
US973425210. Sept. 201215. Aug. 2017Wolfram Alpha LlcMethod and system for analyzing data using a query answering system
US974036919. Dez. 201322. Aug. 2017Palantir Technologies Inc.Systems and methods for providing a tagging interface for external content
US97605561. Apr. 201612. Sept. 2017Palantir Technologies Inc.Systems and methods for annotating and linking electronic documents
US97671722. Okt. 201519. Sept. 2017Palantir Technologies Inc.Data aggregation and analysis system
US977952528. Dez. 20163. Okt. 2017Palantir Technologies Inc.Generating object time series from data objects
US978531727. März 201410. Okt. 2017Palantir Technologies Inc.Presentation and analysis of user interaction data
US97853286. Okt. 201410. Okt. 2017Palantir Technologies Inc.Presentation of multivariate data on a graphical user interface of a computing system
US978577325. März 201510. Okt. 2017Palantir Technologies Inc.Malware data item analysis
US97920204. Mai 201617. Okt. 2017Palantir Technologies Inc.Systems for collecting, aggregating, and storing data, generating interactive user interfaces for analyzing data, and generating alerts based upon collected data
US981756329. Dez. 201414. Nov. 2017Palantir Technologies Inc.System and method of generating data points from one or more data stores of data items for chart creation and manipulation
US20070271226 *19. Mai 200622. Nov. 2007Microsoft CorporationAnnotation by Search
US20080066052 *7. Sept. 200713. März 2008Stephen WolframMethods and systems for determining a formula
US20080077583 *21. Sept. 200727. März 2008Pluggd Inc.Visual interface for identifying positions of interest within a sequentially ordered information encoding
US20100042610 *15. Aug. 200818. Febr. 2010Microsoft CorporationRank documents based on popularity of key metadata
US20100179948 *11. Jan. 201015. Juli 2010Alibaba Group Holding LimitedMethod and system for querying information
US20110035699 *21. Okt. 201010. Febr. 2011Julien RobertBrowsing or Searching User Interfaces and Other Aspects
US20110040752 *14. Aug. 200917. Febr. 2011Microsoft CorporationUsing categorical metadata to rank search results
US20110055759 *8. Nov. 20103. März 2011Julien RobertBrowsing or Searching User Interfaces and Other Aspects
US20110072047 *21. Sept. 200924. März 2011Microsoft CorporationInterest Learning from an Image Collection for Advertising
US20110153582 *22. Dez. 200923. Juni 2011Daniel BuchmannHandling of classification data by a search engine
US20110270819 *30. Apr. 20103. Nov. 2011Microsoft CorporationContext-aware query classification
US20120011112 *6. Juli 201012. Jan. 2012Yahoo! Inc.Ranking specialization for a search
US20120278337 *26. Sept. 20111. Nov. 2012Limelight Networks, Inc.Methods and systems for generating automated tags for video files
US20140100970 *9. Okt. 201310. Apr. 2014Double Verify Inc.Automated Monitoring and Verification of Internet Based Advertising
US20140180692 *19. Febr. 201426. Juni 2014Nuance Communications, Inc.Intent mining via analysis of utterances
US20140188889 *18. Febr. 20133. Juli 2014Motorola Mobility LlcPredictive Selection and Parallel Execution of Applications and Services
US20140379689 *15. Sept. 201425. Dez. 2014Google Inc.Framework for Suggesting Search Terms
US20150026152 *16. Juli 201322. Jan. 2015Xerox CorporationSystems and methods of web crawling
US20150161176 *29. Dez. 200911. Juni 2015Google Inc.Query Categorization Based on Image Results
US20150227583 *20. Apr. 201513. Aug. 2015Google Inc.Managing search results
US20150317365 *30. Apr. 20145. Nov. 2015Yahoo! Inc.Modular search object framework
US20160232246 *16. Nov. 201511. Aug. 2016Sackett Solutions & Innovations, LLCSystem for Search and Customized Information Updating of New Patents and Research, and Evaluation of New Research Projects' and Current Patents' Potential
US20170031927 *30. Juni 20162. Febr. 2017Groupon, Inc.Multi-term query subsumption for document classification
US20170116259 *4. Jan. 201727. Apr. 2017Palantir Technologies Inc.Interactive user interface for dynamic data analysis exploration and query processing
CN102317943A *29. Juli 201111. Jan. 2012华为技术有限公司Method and device for full-text search
WO2012092196A1 *23. Dez. 20115. Juli 2012Alibaba Group Holding LimitedRecommendation of search keywords based on indication of user intention
WO2012106941A1 *29. Juli 201116. Aug. 2012Huawei Technologies Co., LtdMethod and device for full-text search
Klassifizierungen
US-Klassifikation707/759, 707/708, 707/769, 707/805
Internationale KlassifikationG06F17/30
UnternehmensklassifikationG06F17/30707
Europäische KlassifikationG06F17/30T4C
Juristische Ereignisse
DatumCodeEreignisBeschreibung
17. März 2014ASAssignment
Owner name: MICROSOFT CORPORATION, WASHINGTON
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:BENNETT, PAUL N.;CHICKERING, DAVID MAXWELL;COLLINS-THOMPSON, KEVYN B.;AND OTHERS;SIGNING DATES FROM 20090506 TO 20090510;REEL/FRAME:032452/0142
9. Dez. 2014ASAssignment
Owner name: MICROSOFT TECHNOLOGY LICENSING, LLC, WASHINGTON
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MICROSOFT CORPORATION;REEL/FRAME:034564/0001
Effective date: 20141014
26. Okt. 2017MAFP
Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551)
Year of fee payment: 4